华东师范大学学报(自然科学版) ›› 2018, Vol. 2018 ›› Issue (4): 184-194.doi: 10.3969/j.issn.1000-5641.2018.04.018

• 河口海岸学 • 上一篇    

利用SOM神经网络研究长江口邻近海域海表温度特征

白玫, 吴辉   

  1. 华东师范大学 河口海岸学国家重点实验室, 上海 200062
  • 收稿日期:2017-09-12 出版日期:2018-07-25 发布日期:2018-07-19
  • 通讯作者: 吴辉,男,副研究员,研究方向为河口与近海动力学.E-mail:hwu@sklec.ecnu.edu.cn E-mail:hwu@sklec.ecnu.edu.cn
  • 作者简介:白玫,女,硕士研究生,从事河口海岸动力研究.E-mail:deanbai@sina.com
  • 基金资助:
    国家自然科学基金(41776101)

Characteristics of sea surface temperature in the Changjiang Estuary and adjacent waters based on a Self-Organizing Map

BAI Mei, WU Hui   

  1. State Key Laboratory of Estuarine and Coastal Research, East China Normal University, Shanghai 200062, China
  • Received:2017-09-12 Online:2018-07-25 Published:2018-07-19

摘要: 长江口及邻近水域的海表面温度(SST)受到多种动力过程的共同作用,呈现出复杂的时空变化规律.本研究基于数值模拟的结果,采用SOM(Self-Organizing Map)自组织神经网络方法,对长江口及邻近海域SST的季节、大小潮和天气等不同时间尺度上的变化规律进行聚类分析.研究结果表明:长江口及邻近海域的SST在冬、夏季节比较稳定,春、秋季节随时间变化较大;长江口南部沿岸的低温带主要是由长江冲淡水冬季向南扩展造成的;近岸地区尤其是舟山群岛附近存在低温区,与浙江沿岸的上升流相关;苏北地区低温带和向东南方向延伸的低温水舌在冬季和春季最为明显.

关键词: 长江口, SOM神经网络, 海表温度

Abstract: Sea surface temperature (SST) in the Changjiang Estuary and adjacent waters is affected by multiple dynamic processes, which show complex spatiotemporal variability. In this study, the SST variability was investigated by analyzing modeled results with a Self-Organizing Map (SOM) method. The SST variations were clustered on seasonal, spring-neap, and intra-tidal timescales. It was found that the SST pattern remains stable during the winter and summer seasons but highly variable during the spring and autumn seasons. A cold water band is present south of the Changjiang River mouth in the winter season, due to the southward along-shore extension of the Changjiang River plume. A cold water patch occurs near the Zhoushan Islands, caused by upwelling in the Zhejiang coastal water. In the Subei coastal water, the cold water patch and tongues were most significant during the winter and spring seasons.

Key words: Changjiang Estuary, SOM neural network, sea surface temperature

中图分类号: